Re: [GSoC][PATCH 2/4] submodule--helper: introduce for_each_submodule_list()

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 22, 2017, 22:37 UTC
Message-ID: <xmqqefs31b73.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <20170821161515.23775-2-pc44800@gmail.com>

Prathamesh Chavan <pc44800@gmail.com> writes:

> -static void init_submodule(const char *path, const char *prefix, int quiet)
> +static void for_each_submodule_list(const struct module_list list,
> +				    submodule_list_func_t fn, void *cb_data)

It may not be wrong per-se, but can't this just be for_each_submodule()?

Your "justification" may be that this makes it clear that you are iterating over module_list and not other kind of group of submodules, but I would say the design of the subsystem is broken if some places use a list of submodules while some other places use an array of submodules to represent a group of submodules. Especially when there is a dedicated type to hold a group of submodules, i.e. struct module-list, that type should be used consistently throughout the subsystem and API, no?

Show 5 quoted lines

>  {
> +	int i;
> +	for (i = 0; i < list.nr; i++)
> +		fn(list.entries[i], cb_data);
> +}

Also, did you really want to pass the structure by value? At least in C, it is more customary to pass these things by pointer, i.e.

	for_each_submodule(struct module_list *list,
			   for_each_submodule_fn fn,
			   void *cb_data)
	{
		for (i = 0; i < list->nr; i++)
			...

Otherwise you'd be making a copy on stack unnecessarily (ok, "const" might hint a smart compiler to turn this inefficient code to pass it by pointer, but I do not think it is a particulary good to rely on such things).